MASTER
- DISCLAIMED
•ftorlL ipamor«l
by an
THE
USE OF A
SMALL ACCELERATOR
AS A
SOURCE
OF 1H-MeV NEUTRONS
FOR
SHIELDING STUDIES*
G.
T.
Chapnan,
G. L.
Morgan^
and J. W.
McConnell
Engineering Physios Division
The
Oak
Ridge National Laboratory,
Oak
Ridge,
TN
By acceptance
of
this article,
the
publisher
or
recipient acknowledges
the
U.S.
Government's right
to
retain
a
nonexclusive, royalty-free
license
in
and
to
any
copyright
covering
the
article.
It
is
important
in
calculating complex
shields 3uoh
as
those proposed
for the
fusion
reactors
to
ascertain that
the
neutron
cross-section data sets used
in the
calculations
are as
accurate
as
possible
and
that
the
calcu
1
ational methods used
to
transport
the
neutrons
are as
reliable
as
practical.
To
assure that both these
criteria
are met, a
project
at the Oak
Ridge
National Laboratory (ORNL)
is
being conducted
in which
a
snail accelerator
is
used
to
provide 11-HeV neutrons
vis the
Ttd.nJ^He
reaction
and an
NE-213 detector
is
used
to
measure
the
neutron
and
gamma-ray
pulse-height spectra
of the
radiations,
transported through and/or created
in
veryl
thick laminated shields
of
stainless steel:
(type
301) and
borated polyethylene.
To
produce
the
neutron flux required,
the
targets
are
made
by
depositing about
1
rag/cm^
of
Til
onto
a
1.27-cm
circular area
of a
0.251-cm thick copper disk.
The
NE-213
detector
is
operated
in
standard,
state-of-the-art electronic circuits.
A
surface-barrier alpha counter
and a
small
NE-213 detector
are
located permanently
at a
distance
of
about
150 cm
from
the
target
to
monitor
the
reaction rate
in the
target.
The
pulse-height data
are
unfolded
to
produce
energy spectra
by
using
the
computer program
FEHD.2 These results
are
then compared almost
immediately with spectra^ obtained using
two-dimensional radiation transport methods
incorporating 53-neutron, 21-gamma-ray
energy-group cross section data derived from
the VITAMIN-C data
set
(ENDf/B).
1
* Laminated
atainless-s
t eel and
borated polyethylene
shields having thicknesses
up to 412
g/cm
2
have been measured.
The Facility
The facility, designed
and
built
for
conducting
the
measurements described here,
is located
in
Bldg. 6025
at
ORNL. Since
this
is
primarily
an
office complex,
it was
necessary
to
provide shielding
to
meet
all
Health Physics requirements
in
addition
to
that needed
to
reduce
the
backgrounds
produced
by
neutron scattering and/or capture
in
the
room.
Fig. 1
shows
a
cut-away
drawing
of the
facility
as
configured
for
attenuation measurements.
The
components
are
shown
in
place
in the
figure.
A
minimum
of
one meter
of
concrete shielding surrounds
each configuration
in all
directions except
the foward direction.
To
reduce
the
return
of neutrons from
the
wall behind
the
detector, iron slabs were placed between
the
detector
and the
wall. This,
in
effect,
simulated making
the
measurements with
the
detector loea'.ed within
the
experimental
shield.
The accelerator target
(see
below)
is
located
in a
cylindrical
can of
iron with
a
7.5-cm wall thickness. This
can
serves
to
modify
the
nominal 14-MeV neutron source
spectrum
to
more nearly represent
the
softer
spectrum from
the
reactor.
In
addition,
the
source
can
with
its
backup shield
of
lithiated paraffin
in the
accelerator
drift-tube port, reduces
the
number
of
neutrons reflected from
the
back wall
of the
room.
As
designed
and
built,
the
facility
allows making statistically good measurements
with experimental shield-
as
thick
as 112
g/cm
2
in
only
3 to t
hours counting time.
TRITIUM TARGET
5CINTILLATOR
.66
cm
HIGH
JBE
ISCRIMINATION
•IOIOGICAL SHIELO
Figure
1.
Artist
Experimental Facility.
OATA STORAGE
N04I3PHA/C0MPUTER
POM0 DISC STORAGE
rendition of the
Target Calibration
For the deuteron energies used in this
work (E < 300 keV) the T(d,n)
4
He reaction is
essentially isotropic in the center-of-raass
system. Therefore, counting the alpha
particles emitted into a known; osl. id angle
gives an accurate determination of the
neutron source strength. Tht only
calibration constant necessary is the solid
angle transformation factor (lab to
center-of-mass) for the reaction angl« at
which the alpha particles are observed. This
transformation factor is dependent on the
deuteron energy and is calculated using the
Kd^l^He cross section as a function of
energy, the stopping power of the target
(TiT),
and the relative molecular fractions
of the incident beam.- •
The initial test of the technique was
carried out using a very low mass target
holder in which the alpha particles were
detected by a surface-barrier detector at an
angle of 165° and a distance of about 30 en.
The detector was protected from scattered
deuterons by a foil of aluminized mylar with
a thickness of 3.8 x 10"
u
cm. Pulse height
mSTKIIUTIIN
Of
THIS DOCUMENT
IS
UMLlMtTttd
J
X
X
distributions observed
in the
detector showed
a very clean peak, well isolated from noise
and free
of
background.
The
source strength
determined
by the
alpha particle detector
was
compared with
a
simultaneous measurement
of
the neutrons
at 0° (Fig. 2)
using
a
calibrated^ ME-213 detector (with suitable
calculation
of the 0°
neutron-solid-angle
transformation factor). Agreement
was
within
3%
which
is
within
the
uncertainties
of 5$ on
neutron detector efficiency
and 3% on
alpha
particle calibration factor
(due
mainly
to
possible variations
in the
molecular
fractions
of the
incident deuteron beam).
The same low-mass holder
was
installed
in
the
target position
in the
concrete
containment shield
to
calibrate
an
NE-213
neutron monitor located
at 0°. The
production target holder
was
then installed.
This target holder used
the
same
alpha
-
particle detector package relocated
to
90°
and at a
distance
of 150 cm. The
source
strength determined
by
this assembly
was
then
compared
to
that determined
by the
calibrated
neutron monitor
and
found
to
agree within
1$.
The solid angle transformation factor
at 90°
is very nearly equal
to
unity
and is
essentially independent
of
deuteron energy
and therefore independent
of the
relative
molecular fractions
in the
deuteron beam.
(X
tO
J
!
O&NL-0WG 80-H059
-
-
1
1.37
;
Mev
,1,
2.75
t
1
I
\
\
1
V.
G£MMfl ENERGY IMevI
Figure
3. The
unfolded response
of the
HE-213
detector
to
gamma rays from
a
small
21
source.
detector
to
both neutrons
and
gamma rays
are
the
data shown
in Fig. 2 and Fig. 3
respectively.
At.t-.enuati
on
Measurements
Figure U shows the data taken with the
shield configurations given in Table 1. The
solid lines indicate the confidence band for
the unfolded results and the points show the
results of the calculations for comparison.
1
The calculations are smoothed with an
energy-dependent Gaussian distribution
comparable to the actual resolution of the
NE-213 detector. The data are compared for
neutrons above about 850 keV and for gamma
rays above 750 keV. The low-energy neutron
response is governed by the dynamic range and
linearity of the detector system and although
a lower energy gamma-ray threshold is
possible, it was not attempted for these
measurements.
6
9 10 12 14
NEUTRON ENERGY IMeV)
Figure
2. The
unfolded response
of the
NE-213 detector to 11-KeV neutrons with the
target in the Iow-mas3 holder.
ppt-.eo tor .3JI& El enhrnnlns
The primary neutron/gamma-ray detector
used
in
this work
was
made
at
OHNL
and
consisted
of 66.1 g of
NE-213 scintillation
liquid contained
in a
cylindrical
cup
made
of
aluminum
(1.32 x 10"
2
cm
wall thickness)
and
coated
on the
inside with titanium dioxide
reflector paint.
The
detector
was
mounted
on
an 8850 phot omultiplier tube which
was
used
with state-of-the-art, pulse-shape
discrimination circuitry
to
distinguish
between neutrons
and
gamma rays.
The
photomultiplier tube base
and the
preamplifier were also designed
and
built
at
ORHL, Indicative
of the
response
of the
Table 1
COMPOSITION
AND
THICKNESS
OF
STAINLESS STEEL
304
AND BORATEO POLYETHYLENE SLABS
SIM*
CONflGU RATION
a
t 15
M
}
3041
1 »«I
MM
1 *i«
t
30
41
•STAIHtESSSTCEl TYPE
COMfOS
TIOK
SS-JH
U* SS-1M IP SIM
SIM
TttttKNtSSUn)
SM
sot
sot
5.M
3M
E.NE
SM
sot
SM
sot
SM
SM
SM
5M
SM
SM
TOTAL
SLA
THICK1ES5
Itml
O
W.M
»«
40
M
«72
5010
»M
Figure
1. The
ganma-ray
and
neutron data
measured with
the
detector
on the
axis
of
symmetry showing
the
effects
of
attenuation
by varous thicknesses
of
stainless steel
and
borated polyethylene.
Streaming Measurements
The facility
may be
reconfigured
to
measure
the
streaming
of
neutrons
a.id
gamma
rays through ducts.
In
this case,
a
duct
of
a given length
and
diameter
is
placed
in the
containment shield
and
surrounded
by
concrete
to eliminate
any
voids.
One
such duct
consic'
ing of a
60.96-cm extension
of the
target
can
described above
has
been measured
and
the
results
are
shown
in Fig. 5.
Again,
the solid l'ines show
the
confidence band
for
the unfolded data
and the
points show
the
calculated results.6
The
calculated results
for
the
detector
off the
axis
of
symmetry
show
an
inconsistancy with
the
measured data.
This
is
tentatively attributed
to the
inability
of the P3
Legendre expansion used
in
the
calculations
to
approximate
the
neutron scattering a.'.gular distribution.
(The data shown here have
the
single-scattered contributions subtracted.)
Further investigations
of
this discrepancy
arj underway
at
ORNL using higher order
expansions
and
Monte Carlo techniques.°
A small facility
at the Oak
Ridge
National Laboratory
has
produced neutron
and
gamma-ray data with sufficiently good
statistics
to
allow
a
direct comparison with
radiation transport calculations based
on
group cross-section data derived from
the
VITAMIN-C data
set. A
close working
association with
the
group performing
the
calculations provides
an
almost immediate
comparison
of the
calculations
and the
experimental data. This
haa
resulted
in a
significant improvement
in the
quality
of
both methods
for
evaluating
the
effectiveness
of
the
proposed fusion-reactor shield.
NEUTRON £.\EHGV 'M.yl
Figure
5. The
neutron data measured
to
show
the effects
of the
radiations streaming
through
a
duet.
The
data show
the
effects
of
moving
the
detector
oft the
axis
of
symmetry.
Tha contributions from single scatterings
have been subtracted from
the
calculated
neutron data
(see
text).
5.
References
•Research sponsored
by the U. S. DOE
Office
of
Fusion Energy under Contract
No.
W-7
>\
0 5 - e ng - 2 6
with Union Carbide
Corporation.
Present address:
Los
Alamos Scientific
Laboratory,
Los
Alamos,
New
Mexico 87545.
W.
R.
Burris
and V. V.
Verbinski,
Nucl. Instrum. Methods
fii, 181
(1969).
R.
T.
Santoro,
R. G.
Alsmiller,
Jr.,
J.
M.
Barnea,
and E. M.
Oblo..
. Int.
Conf.
Nuclear Cross Sections
for
Technology, Knoxville, Tennessee, October,
1979.
ORNL/RSIC-37, Radiation Shielding
Information Center,
Oak
Ridge National
Laboratory (1975) DLC/H1-VITAMIN-C.
V.
V.
Verbinski,
W. R.
Burris,
T. A.
Love,
W.
Zobel,
N.
Inst.
and
Meth.,
6JL,
R.
T.
Santoro,
R.
J.
M.
Barnes
and G.
Nucl,
W.
Hill,
8(1968)
G. Alsmiiler, Jr.
,
T. Chapman,
to be
published
in the
Proceedings
oX the
Fourth
A2L2. Tnn<
oal
M_£e_lin£. OJtt iilfi Technology
of
£.aaicall£i
RU.S.L&AC
Eualaa^. King
of
Prussia,
PA, 1980.
